Thanks for checking out my work. All of my papers should be available from my Google Scholar profile (I don't put them here because publishers don't like that).
Click on the links below or scroll down to learn more about the various topics I'm interested in.
User Studies in AI
Measuring machine performance is usually done by calculating the error between output and some ground truth or some other distance between performance and ideal. However, this formulation of performance ignores that machines were designed for humans to use. I have designed studies to show that human perception of GenAI output realism does not align with mathematical evaluations of model quality, and I have devised a method to evaluate XAI interpretability for neuronal processing visualizations. These studies are a reminder that the usage of AI must always be considered next to the basic research.
Stein, G., …, Taylor, J.E.T., & Loaiza-Ganem, G. (2023). Exposing flaws of generative model
evaluation metrics and their unfair treatment of diffusion models. NeurIPS 2023, New Orleans, LA, USA.
Dey, N.S., Taylor, J.E.T., Tripp, B.P., Wong, A., & Taylor, G.W. (2020). Identifying and interpreting tuning dimensions in deep networks. 2nd Annual for Shared Visual Representations in Human and Machine Intelligence at NeurIPS 2020, Vancouver, BC, Canada.
Artificial Cognition
Artificial intelligence powered by deep learning has produced algorithms of incredible predictive power. As these machines are entrustred to make decisions that affect our safety and well-being, it is vitally important that we understand how these black boxes operate. Our solution is to examine AI the way we investigate another black box - the human mind. We think research methods from cognitive psychology can help us test falsifiable hypothesis and develop a working theory of behaviour for these inscrutable machines.
Taylor, J.E.T. & Taylor, G.W. (2020). Artificial Cognition: How Experimental Psychology can Help Develop Explainable Artificial Intelligence. Psychonomic Bulleting & Review.
Taylor, J.E.T., Shekhar, S., & Taylor, G.W. (2020). Response Time Analysis for Explainability of Visual Processing in CNNs. Minds vs. Machines Workshop at CVPR 2020, Seattle, WA, USA.
Human Visual Cognition
There's a lot of information out there in the world. Too much to process, in fact. Attention is the handy filter that picks out the good stuff so that you can ignore the noise. The factors that determine where and how attention is deployed form the cornerstone of more complex behaviours, such as action. However, we're still struggling to understand even the most basic principles of human attention. For example, if you want to understand how a human (or a self-driving car!) avoids obstacles, you first need to know how the visual system picks up suddenly transforming information on the road, or how attention separates objects from the environment. By understanding the principles of attention at a basic level, we can appreciate how we accomplish more difficult tasks.
Taylor, J. E. T., Hilchey, M. D., Weidler, B. J., & Pratt, J. (2022). Eliminating the
low-prevalence effect in visual search with a remarkably simple strategy. Psychological
Science, 33(5).
Taylor, J. E. T., Hilchey, M. D., & Pratt, J. (2018). Out with the new, in with the old: Exogenous orienting to locations with physically constant stimulation. Psychonomic Bulletin & Review.
Taylor, J. E. T., Chan, D., Bennett, P. J., & Pratt, J. (2015). Attentional cartography: Mapping the distribution of attention across space and time. Attention, Perception, & Psychophysics, 77(7), 2240-2246.
Taylor, J. E. T., Rajsic, J. & Pratt, J. (2016). Object-based selection is contingent on attentional control settings. Attention, Perception, & Psychophysics, 78(4), 988-995.
Traditional approaches to the study of perception typically assume that perception is modular, which is to say that it's not affected by the way you think or act. Contrary to this notion, part of my research investigates the manner in which the ability to act modulates our perception of the environment. For example, underwater distances appear shorter when wearing flippers (effort-based scaling), and walls appear shorter when you can climb them (skill-based scaling).
Taylor, J. E. T., Witt, J. K., & Sugovic, M. (2011). When walls are no longer barriers: Perception of wall height in parkour. Perception, 40, 757-760.
Witt, J. K., Schuck, D. M., & Taylor, J. E. T. (2011). Action-specific effects underwater. Perception, 40, 530-537.
Witt, J. K., Sugovic, M., & Taylor, J. E. T. (2011). Action-specific effects in a social context: Others’ abilities influence perceived speed. Journal of Experimental Psychology: Human Performance & Perception, 38, 715-725.
One of the biggest problems our perceptual systems need to solve is how to isolate and represent the action-relevant information in the environment. For example, attention has built-in priorities for attending to things in the space near the hands, and the information that we attend to changes depending on what tools we're using.
Taylor, J. E. T. & Witt, J. K. (2014). Altered attention for stimuli on the hands. Cognition, 133, 211-225.
Taylor, J. E. T., Gozli, D. G., Chan, D., Huffman, G., & Pratt, J. (2015). A touchy subject: Advancing the modulated visual pathways account of altered vision near the hands. Translational Neuroscience, 6(1), 1-7.
Taylor, J. E. T., Pratt, J., & Witt, J. K. (2015). Joint attention for stimuli on the hands: Ownership matters. Frontiers in Psychology, 6(543).
Pratt, J., Taylor, J. E. T., & Gozli, D. (2015). Action and attention. Chapter in Fawcett, J., Risko, E. F., & Kingstone, A. (Eds.), Handbook of Attention. Boston, USA: MIT Press.
Taylor, J. E. T., Pratt, J., & Witt, J. K. (2017). A different kind of weapon focus: Sensorimotor experience reduces change blindness. Cognitive Research: Practical Implications, 2(1), 1-9.
Observing the actions of others implicitly activates a corresponding representation in the motor system of the observer, a phenomenon known as motor resonance. My research show that this sympathetic motor activation develops with practice in the perceived action. Interestingly, these simulations can occur simply by observing the 'trace' of an action. For example, piano players demonstrate this motor resonance when they hear a recording of piano performance, and most people demonstrate motor resonance when they observe a painting with evocative brushstrokes.
Taylor, J. E. T., Witt, J. K., & Grimaldi, P. J. (2012). Uncovering the connection between artist and audience: Viewing painted brushstrokes evokes corresponding action representations in the observer. Cognition, 125, 26-36.
Taylor, J. E. T & Witt, J. K. (2014). Listening to music primes space: Pianists, but not novices, simulate heard actions. Psychological Research, 79(2), 175-182.